Hydrosilylation reaction which is addition reaction of a Si—H functional compound to a compound having a carbon-carbon double bond or triple bond is a useful means for the synthesis of organosilicon compounds and an industrially important synthesis reaction.
As the catalyst for hydrosilylation reaction, Pt, Pd and Rh compounds are known. Among others, Pt compounds as typified by Speier catalyst and Karstedt catalyst are most commonly used.
While several problems arise with reaction in the presence of Pt compounds as the catalyst, one problem is that upon addition of a Si—H functional compound to terminal olefin, a side reaction due to internal rearrangement of olefin takes place. Since this system does not exert addition reactivity to the internal olefin, unreacted olefin is left in the addition product. To drive the reaction to completion, it is necessary to use an excess amount of olefin in advance by taking into account the fraction left as a result of side reaction.
Another problem is that the selectivity of α- and β-adducts is low depending on the type of olefin.
The most serious problem is that all the center metals Pt, Pd and Rh are quite expensive noble metal elements. As metal compound catalysts which can be used at lower cost are desired, a number of research works have been made thereon.
For example, reaction in the presence of iron-carbonyl complexes (Fe(CO)5, Fe2(CO)12) is known from Non-Patent Document 1, although this reaction requires reaction conditions including as high a temperature as 160° C. or light irradiation (Non-Patent Document 2).
For these iron-carbonyl complexes, it is reported in Non-Patent Document 3 and Patent Document 1 that dehydrogenation silylated products are obtained rather than the addition reaction.
Also Non-Patent Document 4 and Patent Document 2 report a reaction of methylvinyldisiloxane and methylhydrogendisiloxane in the presence of an iron-carbonyl complex coordinated with a cyclopentadienyl group. Since dehydrogenation silylation reaction takes place along with the relevant reaction, the selectivity of addition reaction is low.
With respect to reaction in the presence of an iron catalyst having a terpyridine ligand (Non-Patent Document 5), a large excess of a reducing agent (NaBHEt3) is necessary as a reaction co-agent. Although PhSiH3 and Ph2SiH2 add to olefins, more useful trialkylsilanes, alkoxysilanes and siloxanes have poor addition reactivity to olefins.
Non-Patent Document 6 reports that from reaction in the presence of an iron catalyst having a terpyridine ligand and a bistrimethylsilylmethyl group, an addition reaction product is obtained in high yields. This method needs some steps until the catalyst is synthesized, including first synthesizing a terpyridine-iron complex as a catalyst precursor and introducing a bistrimethylsilylmethyl group therein at a low temperature, which steps are not easy industrially.
Also, Non-Patent Documents 7 and 8 report iron complexes having a bisiminopyridine ligand. It is disclosed that they exhibit high reactivity to alkoxysilanes and siloxanes under mild conditions.
The reaction using the complex, however, suffers from several problems including low reactivity with internal olefin, the use of sodium amalgam consisting of water-prohibitive sodium and highly toxic mercury and requiring careful handling (or use of water-prohibitive NaBEt3H) for complex synthesis, low stability of the complex compound itself, a need for a special equipment like a glove box for handling, and a need for storage in an inert gas nitrogen atmosphere at low temperature.
Non-Patent Documents 9 to 14 report examples of reaction in the presence of cobalt-carbonyl complexes (e.g., Co2(CO)5), but they are unsatisfactory in reaction yield and reaction molar ratio. No reference is made to addition reactivity to siloxanes.
Also an example of reaction of olefin with trialkylsilane in the presence of a cobalt-carbonyl complex substituted with a trialkylsilyl group is reported in Non-Patent Document 15, but the yield is low and the selectivity is low.
Non-Patent Document 16 reports reaction of olefin with trialkylsilane in the presence of a cobalt-phosphite complex coordinated with a cyclopentadienyl group, and Non-Patent Document 17 reports reaction of olefin with trihydrophenylsilane in the presence of a cobalt complex coordinated with N-heterocyclocarbene. Because of low stability, these complex compounds require a special equipment like a glove box for handling and an inert gas atmosphere and a low temperature for storage.
Also Patent Documents 3 to 6 report iron, cobalt and nickel catalysts having terpyridine, bisiminopyridine and bisaminoquinoline ligands. Like the above-cited Non-Patent Documents 6 to 8, there are problems including industrial difficulty of synthesis of a catalyst precursor or synthesis of the complex catalyst from the precursor, low stability of the complex compound itself, and a need for a special equipment for handling.
Patent Document 7 discloses a method of conducting reaction in the presence of a complex catalyst having a bisaminoquinoline ligand, using Mg(butadiene).2THF or NaEt3BH as the catalyst activator. There are the same problems as above and the yield of the desired product is less than satisfactory.
Many examples of the nickel complex catalyst are reported. For example, a catalyst having a phosphine ligand (Non-Patent Document 18) lacks in selectivity and requires careful handling and storage.
With a vinylsiloxane-coordinated catalyst (Non-Patent Document 19), a dehydrogenation silylated product becomes predominant, indicating low selectivity of addition reaction.
With an alkylphosphine-coordinated catalyst (Non-Patent Document 20), the yield is low, and trihydrophenylsilane is not a substrate of industrial worth.
A bisamide-bearing catalyst (Non-Patent Document 21) needs careful handling and storage, and dihydrodiphenylsilane is not a substrate of industrial worth.
A catalyst having N-heterocyclocarbene ligand (Non-Patent Document 22) has low selectivity of reaction, and trihydrophenylsilane is not of industrial worth.
Many rhodium complex catalysts are reported. For example, catalysts having a carbonyl or cyclooctadienyl (COD) group and a N-heterocarbene ligand (Non-Patent Documents 23, 24) require handling and storage in an inert gas atmosphere because the complex compounds have low stability.
Non-Patent Document 25 discloses to conduct reaction in the presence of an ionic liquid in order to enhance reactivity. The step of separating the ionic liquid from the reaction product is necessary. Since the catalyst used therein has a COD group and a N-heterocarbene group as the ligand, the same problems as described above are left.
Also Non-Patent Document 26 reports an exemplary catalyst which allows for preferential progress of dehydrogenation silylation reaction.
Furthermore, Non-Patent Document 27 reports an example in which an isocyanide compound is added to a complex catalyst to form a catalyst, which is used in hydrosilylation reaction without isolation. A study on reactivity with three types of silanes shows that the order of reactivity is from dimethylphenylsilane, which gives the highest yield (yield 81%), next triethylsilane (yield 66%), to triethoxysilane (yield 40%). The reactivity with triethoxysilane which is of the most industrial worth among the three types of silanes is not so high, while the reactivity with siloxanes is reported nowhere.
In addition, the precursor catalyst having a COD group as the ligand requires careful handling and storage.
On the other hand, Non-Patent Document 28 reports that a rhodium catalyst having an acetylacetonato or acetate group enables addition reaction of triethoxysilane in high yields.
Although this method has the advantage of easy storage and handling of the catalyst, no study is made on reactivity with siloxanes which are more useful from the industrial standpoint.
In addition, rhodium is likewise an expensive noble metal element. Its catalytic function must be further increased to a higher activity before it can be used in practice as a platinum replacement.
The catalysts with their application to organopolysiloxanes being borne in mind include a catalyst having a phosphine ligand (Patent Document 8), a catalyst having an aryl-alkyl-triazenide group (Patent Document 9), a colloidal catalyst (Patent Document 10), a catalyst coordinated with a sulfide group (Patent Document 11), and a catalyst coordinated with an amino, phosphino or sulfide group and an organosiloxane group (Patent Document 12).
However, reactivity is empirically demonstrated with respect to only platinum, palladium, rhodium and iridium which are expensive metal elements. Thus the method is not regarded cost effective.
In Examples of Patent Documents 13 and 14, only well-known platinum catalysts are demonstrated to exert a catalytic effect while the structure which is combined with another metal to exert catalytic activity is indicated nowhere.
Patent Documents 15 to 17 disclose catalysts coordinated with carbene. Patent Document 15 does not discuss whether or not the catalyst is effective to hydrosilylation reaction.
Patent Documents 16 and 17 disclose catalysts coordinated with carbene and vinylsiloxane, but describe only platinum catalysts in Examples.
In addition, the metal catalysts coordinated with carbene require careful handling because the complex compounds have low storage stability.
Patent Documents 18 and 19 disclose ruthenium catalysts coordinated with η6-arene or η6-triene. These catalysts have inferior reactivity to platinum catalysts and require careful handling because the complex compounds have low storage stability.
Patent Documents 20 to 26 disclose a method of mixing a metal salt with a compound which coordinates to the metal and using the product as a catalyst rather than the use of metal complexes as the catalyst. Although these Patent Documents describe the progress of hydrosilylation with several exemplary combinations, the yield and other data are described nowhere, and the extent to which the reaction takes place is not evident. Although all Examples use ionic salts and hydride reducing agents as the activator, catalytic activity is not detected in most Examples.